Stories vs. Statistics

Air travel plunged after September 11th 2001. Which is understandable. Everyone saw videos of the planes hitting the towers. News stations put the clips on repeat for months. Add to it the consensus that a follow-up attack was a matter of when, not if, and people felt reckless approaching airplanes. Americans flew 7% fewer miles in all of 2002 than they did in the first nine months of 2001.

But people still had places to go. So they drove. Miles driven by car rose 4% in 2002 despite a lingering recession, with long-distance rural highway travel rising the most – an indication that people were driving more specifically because they were flying less.

Which was a problem, because driving is far deadlier than flying.

More Americans die in car accidents every 13 days than have died in all commercial plane accidents in the last 27 years combined, according to data from the Department of Transportation. About 1.5 Americans die in car accidents per 100 million miles traveled. For commercial airlines the average number is 0.17, and in many years is literally 0.00. If a 9/11 happened every year, flying would still be safer than driving was in the 1950s.

In 2004, German psychologist Gerd Gigerenzer calculated the off-trend rise in car travel after 9/11 and its correlation to a rise in auto fatalities. His conclusion: “The number of Americans who lost their lives on the road by avoiding the risk of flying was higher than the total number of passengers killed on the four fatal flights.”

We’re in the era of Big Data, where we have statistics on everything.

But the cult of data often underestimates that stories are more powerful than statistics.

You can show people statistics about how driving is more dangerous than flying. But it’s hard to contextualize a number. A Canadian survey once showed that almost half of adults couldn’t correctly answer the question, “How many millions are in a trillion?” So even simple statistics seem like a foreign language, where you know a few words but can’t turn them into anything useful. But show people video of jets taking down iconic buildings, or pictures of grieving families, and there’s no nuance. The story is instantly persuasive.

Statistics are useful to the extent that we find their insights more persuasive than anecdotal stories.

But we often don’t.

Statistically, a company should want some of its new projects to fail. Realistically, admitting failure burns so badly that many companies never try new stuff.

Statistically, a venture portfolio can lose on most of its investments while still doing well overall. Realistically, sinking time and money into a company that goes under can turn you into once bitten, twice shy.

Statistically, it’s inevitable and OK if a few companies in the S&P 500 are failures. Realistically, watching Enron’s CEO hauled off in handcuffs shakes your faith in capitalism.

The gap between what works in a spreadsheet and what’s practical in real life can be a mile wide. This usually isn’t because we don’t know the statistics. It’s because real-life stories are so effective at showing us what certain parts of a statistic mean that we extrapolate them into something more powerful than they should be.

Daniel Kahneman once wrote:

A salient event that attracts your attention will be easily retrieved from memory. Divorces among Hollywood celebrities and sex scandals among politicians attract much attention, and instances will come easily to mind. You are therefore likely to exaggerate the frequency of both Hollywood divorces and political sex scandals.

This idea has been amplified in recent years by phone cameras, blogs, Twitter, and Facebook, which give stories and anecdotes a microphone like they’ve never had. While Big Data has grown, Big Anecdotes has exploded. Since the latter can be more persuasive, we’re in a weird spot where we have more statistics than ever to help us make smart decisions, but it’s also easier than ever to let stories pull us toward bad decisions.

It’s why active stock investing is more popular than it should be. Understanding the math about the difficulty of beating the market takes effort. But recalling Warren Buffett’s success story is instantly understandable.

It’s why young college grads rush into glamours-sounding careers despite evidence of future burnout and career dissatisfaction.

And it’s why companies often struggle to bridge the gap between a cool product and a viable business. “I love this product” is a story product managers can latch on to. “I need this product, have room in my household budget to pay for it, and will continue paying for it in the future at a price that covers your cost of capital,” requires bland statistics.

Gerd Gigerenzer writes in his book Risk Savvy:

People aren’t stupid. The problem is that our educational system has an amazing blind spot concerning risk literacy. We teach our children the mathematics of certainty— geometry and trigonometry— but not the mathematics of uncertainty, statistical thinking. And we teach our children biology but not the psychology that shapes their fears and desires. Even experts, shockingly, are not trained how to communicate risks to the public in an understandable way.

Hard to argue. We’d all be better off with a deeper grasp of the statistics that explain our businesses, investments, and careers.

But if we ignore how easily statistics are kidnapped and brainwashed by stories, Big Data won’t do us much good. Or at least it won’t fulfill its potential.

Many – I’d say most – of the world’s greatest investors find their edge in psychology and understanding human behavior, rather than an informational advantage. They have statistics, and use them wisely. But they’re also masters at understanding the power and persuasion of stories, both on themselves and other people.

We need more of this kind of thinking in all fields. More acceptance that understanding the psychology of how people actually act is just as, if not more, important than the data that’s supposed to guide how we should act.

If statistics have a dark side, it’s the assumption that they’re useful on their own, rather than as a tool to help guide gullible and story-addicted people. Since new statistics are fighting against millions of years of primal instincts, conquering the power of stories will always be harder and more rewarding than uncovering new numbers.